Advocacy & Safety - A car that drives itself...

I have mentioned autonomous drive vehicles here before, a couple of times, and apparently to the dismay of some motor vehicle loving cyclists. Such vehicles are just around the corner... according to the article below:

http://news.yahoo.com/s/ap/20080107/ap_on_hi_te/gadget_show_driverless_cars


http://d.yimg.com/us.yimg.com/p/ap/20080106/capt.eef8e207f2154eb8bc73e12b5ed8a032.gadget_show_driverless_cars_dt103.jpg?x=400&y=266&sig=CNH1b046Zj4pEhZODKeq8Q--

DETROIT - Cars that drive themselves — even parking at their destination — could be ready for sale within a decade, General Motors Corp. executives say.

GM, parts suppliers, university engineers and other automakers all are working on vehicles that could revolutionize short- and long-distance travel. And Tuesday at the Consumer Electronics Show in Las Vegas GM Chief Executive Rick Wagoner will devote part of his speech to the driverless vehicles.

"This is not science fiction," Larry Burns, GM's vice president for research and development, said in a recent interview.

The most significant obstacles facing the vehicles could be human rather than technical: government regulation, liability laws, privacy concerns and people's passion for the automobile and the control it gives them.

Much of the technology already exists for vehicles to take the wheel: radar-based cruise control, motion sensors, lane-change warning devices, electronic stability control and satellite-based digital mapping. And automated vehicles could dramatically improve life on the road, reducing crashes and congestion.

If people are interested.

"Now the question is what does society want to do with it?" Burns said. "You're looking at these issues of congestion, safety, energy and emissions. Technically there should be no reason why we can't transfer to a totally different world."

GM plans to use an inexpensive computer chip and an antenna to link vehicles equipped with driverless technologies. The first use likely would be on highways; people would have the option to choose a driverless mode while they still would control the vehicle on local streets, Burns said.

He said the company plans to test driverless car technology by 2015 and have cars on the road around 2018.

Sebastian Thrun, co-leader of the Stanford University team that finished second among six teams completing a 60-mile Pentagon-sponsored race of driverless cars in November, said GM's goal is technically attainable. But he said he wasn't confident cars would appear in showrooms within a decade.

"There's some very fundamental, basic regulations in the way of that vision in many countries," said Thrun, a professor of computer science and electrical engineering.

The Defense Department contest, which initially involved 35 teams, showed the technology isn't ready for prime time. One team was eliminated after its vehicle nearly charged into a building, while another vehicle mysteriously pulled into a house's carport and parked itself.

Thrun said a key benefit of the technology eventually will be safer roads and reducing the roughly 42,000 U.S. traffic deaths that occur annually — 95 percent of which he said are caused by human mistakes.

"We might be able to cut those numbers down by a factor of 50 percent," Thrun said. "Just imagine all the funerals that won't take place."

Other challenges include updating vehicle codes and figuring out who would be liable in a crash and how to cope with blown tires or obstacles in the road. But the systems could be developed to tell motorists about road conditions, warn of crashes or stopped vehicles ahead and prevent collisions in intersections.

Later versions of driverless technology could reduce jams by directing vehicles to space themselves close together, almost as if they were cars in a train, and maximize the use of space on a freeway, he said.

"It will really change society, very much like the transition from a horse to a car," Thrun said.

The U.S. government has pushed technology to help drivers avoid crashes, most notably electronic stability controls that help prevent rollovers. The systems are required on new passenger vehicles starting with the 2012 model year.

Vehicle-to-vehicle communication and technology allowing cars to talk with highway systems could come next.

Still in debate are how to address drivers' privacy, whether current vehicles can be retrofitted and how many vehicles would be need the systems to develop an effective network.

"Where it shakes out remains to be seen but there is no question we see a lot of potential there," said Rae Tyson, a spokesman for the National Highway Traffic Safety Administration.

*******************************************************

Now the thing we cyclists have to do is ensure that any such vehicle/system, takes us into account. On the plus side such an autonomous vehicle is not likely to break the law, nor act aggressively toward cyclists... on the flip side we must ensure, through pressure to law makers, that such vehicles do not impede our safety as legitimate road users.

Various levels of automated driver assist technologies are coming online, and the self-piloting automobile (true AUTOmobile, I suppose??) is the obvious ultimate goal, once the equally obvious technical reliability and legal liability issue are overcome. One big motivation is to increase the capacity of freeways by platooning cars within automated lanes, a concept conservative "5 second gap" and "bubble" motorists such as I find terrifying.

1. Maximum speed 25 mph
2. Cheap, readily available shut down remotes (think garage down opener so that the car could be shut down if anyone thought it was operating poorly.

Then I'd consider them.

But can a robot car lean on the horn and sideswipe a cyclist as well as a human driver? I think not.

1. Maximum speed 25 mph
2. Cheap, readily available shut down remotes (think garage down opener so that the car could be shut down if anyone thought it was operating poorly.

Then I'd consider them.

The interesting thing is that the government is working to mandate them... so motorists may eventually not have a choice... unless they plan on driving on closed tracks.

From the article I posted is this little tidbit: "The U.S. government has pushed technology to help drivers avoid crashes, most notably electronic stability controls that help prevent rollovers. The systems are required on new passenger vehicles starting with the 2012 model year."

Remember the government pushed for seatbelts, airbags, collapsing steering wheel columns, and a host of other things for the safety of the auto users. The nanny state will ensure that your future auto is safe for you, whether you want it or not.

This is an idea that's actually very old. Part of the original ISTEA plan depended on self driving cars on most major roads by 2030 or something like that.

I used to think it was a bad idea, but people are getting more and more stupid every day. And it shows in thier driving. He!!, even Microsoft could probably write a software program that drives better than 50% of the drivers on the road right now.

Az

For a long time, I thought the idea of a computer driving a car was just a Bad Idea, maybe even a frightening idea. The humans were just so much better at the tasks required to drive a car.

From a lot of what I've seen lately, I'm not so sure, and that's more due to changes I've seen in the human drivers than changes I've seen in computer technology.

"DETROIT - Cars that drive themselves — even parking at their destination — could be ready for sale within a decade, General Motors Corp. executives say."

Ha! These are already on our highways! Surly one in ten cars out there is driving it's self while the motorist is busy texting! ;)

Ha! These are already on our highways! Surly one in ten cars out there is driving it's self while the motorist is busy texting! ;)

Right... isn't that what cruise control is for... :rolleyes:

Flawed as they are, humans can make judgement calls, like hit the truck instead of the pedestrian. Or run off the road instead of hitting the pedestrian.

I can see it now, plug in modules that allow you to reprogram your auto drive for faster speeds, faster cornering, or whatever.

People are already hacking GPS route finding systems, just wait utnil the hack auto drive cars.
******* will select their victims and have the victims' own car deliver them. Ex-husbands will program the ex-wifes' car to drive into a lake.

1. Maximum speed 25 mph
2. Cheap, readily available shut down remotes (think garage down opener so that the car could be shut down if anyone thought it was operating poorly.

Then I'd consider them.
This attitude always interested me, being as computers make considerably fewer mistakes than humans, and can scan using pure numerical data on the world around them with millisecond response time, and don't get distracted. (To be exact, they don't make any mistakes, they either break or are programmed wrong)

Me, I just LIKE driving, it's the only thing I fear... I don't get to drive anymore, the car does it for me. As for safety, I realize the practicality of using computers that error much less frequently than humans is a smart idea.

Also, considering that throttle, steering, brakes and traction control is already mainly computer controlled in cars now a day, I don't see what the huge difference with them filling in the rest of the gap. Mind you, these systems have no easy override, and malfunctions can end in death, however we practically never see them.


People are already hacking GPS route finding systems, just wait utnil the hack auto drive cars.
******* will select their victims and have the victims' own car deliver them. Ex-husbands will program the ex-wifes' car to drive into a lake.
I'm pretty sure driving systems will be most likely ROM without the ability to wirelessly hijack it for the obvious reason you already mentioned. It would be stupid to let an outside device influence a machine that should be self-sufficient on it's own programming.

Sure, if someone rips out your car's computer they could technically hijack that, but already they can hijack your electronic throttle system to kill you in the same way.


Flawed as they are, humans can make judgement calls, like hit the truck instead of the pedestrian. Or run off the road instead of hitting the pedestrian.

I can see it now, plug in modules that allow you to reprogram your auto drive for faster speeds, faster cornering, or whatever.
Computers can be programmed to make judgment calls like those, they're nothing but logic, the system should be able to pick up the size of an object, and distance from any objects within 360 degrees of the car depending on sensors. Way better than a human ever could, quicker, with greater precision, and 100% optimal usage of every aspect of the car's handling abilities.

However, due to the system itself, accidents like this would be so extremely rare it's laughable. A computer can track all traffic on the road in visible range without networking (and even invisible traffic [due to obstacles] if they were networked), therefore there are no surprises, every car knows what every other car in the vicinity is doing through various sensors.

Ha! These are already on our highways! Surly one in ten cars out there is driving it's self while the motorist is busy texting! ;)

Sad but true. The United States automobile industry decided a long time ago that computer driven cars were the wave of the future. In order to break us in to the concept, cars today are full of all kinds of gizmos, GPS, widescreen video, etc. to get is used to the idea of not having to pay attention to driving, with the unintended consequence of everyone talking on their cell phones or taxing, instead of pay attention to driving.

The American car of the future, which is already partially here, is just a rolling entertainment center, measured by the number of cup holders and TV screens the car has, with as little input from the driver is possible.

What a difference in Europe, where high-performance cars like Porsche don't even have cup holders. Europeans understand the concept that cards are meant to be driven, not entertain you.

[QUOTE=StrangeWill;5944365]


I'm pretty sure driving systems will be most likely ROM without the ability to wirelessly hijack it for the obvious reason you already mentioned. It would be stupid to let an outside device influence a machine that should be self-sufficient on it's own programming.
QUOTE]

SO you're telling me that a remote shut down won't be integrated into these so that if they go crazy, as computers often do, you can't shut them down? Or they won't have remote shut down for police use?

Of course they will.

If these self-driving cars work as "well" as GPSs we have nothing to fear. Honestly, that's an invention specifically for a man to drive a woman crazy. Gotta bite your tongue the whole time. I know we have to go that way but the darn GPS is sending us on a wild goose chase and I gotta just be quiet or gadget-guy will really get mad because you can't question the almighty gadget.

Self-driving cars will drive you in circles as you look out the window muttering helplessly, "but I want to go over there."

Here's what I wonder about: very few drivers obey all traffic laws. Most speed; about 9 in 10 fail to stop completely for stop signs or right on red; almost nobody yields to pedestrians in crosswalks; I could go on. So will these devices be programmed to drive the way that people really drive, or in strict accordance with the law? I could see the first option being unpopular and the second exposing the manufacturer to liability.

Then there's the cyclist's perspective. How will these things be programmed to drive around cyclists? In many states it's illegal to pass a cyclist on a double yellow. Will they just wait if there's a cyclist in the road? Will they honk? That's illegal too.

Or will we see a wholesale revision of the traffic laws? One, most likely, that would further marginalize non-motorized vehicles, and even non-automated ones.

Newspaper articles make it seem like cars are already driving themselves:

"Pedestrian killed by SUV"

"Cyclist hit by SUV"

Or, my favorite,
"SUV carrying drunk driver hits cyclist"

Here's what I wonder about: very few drivers obey all traffic laws. Most speed; about 9 in 10 fail to stop completely for stop signs or right on red; almost nobody yields to pedestrians in crosswalks; I could go on. So will these devices be programmed to drive the way that people really drive, or in strict accordance with the law? I could see the first option being unpopular and the second exposing the manufacturer to liability.

Then there's the cyclist's perspective. How will these things be programmed to drive around cyclists? In many states it's illegal to pass a cyclist on a double yellow. Will they just wait if there's a cyclist in the road? Will they honk? That's illegal too.

Or will we see a wholesale revision of the traffic laws? One, most likely, that would further marginalize non-motorized vehicles, and even non-automated ones.

Good questions... but can you imagine all cars suddenly obeying all the rules of the road... boggles the mind doesn't it?

SO you're telling me that a remote shut down won't be integrated into these so that if they go crazy, as computers often do, you can't shut them down? Or they won't have remote shut down for police use?

Of course they will.
"As computers often do"
What?! Paranoid beyond reality much?

Yes, there should be a shutdown for mainly other reasons (if your car goes sideways there is no way in hell that you're going to recover it any better than a computer that can independently control all 4 tires and sense every minor movement in the vehicle), but you do realize that most commercial and military aircraft in the air are LARGELY computer controlled. Along with most of the driving systems in cars now a day anyway. ABS, TCS, DBW and Transmission control, all which keep you on the road much better than without.

If anything human intervention will pollute the system and make it thousands of times more inefficient.


Here's what I wonder about: very few drivers obey all traffic laws. Most speed; about 9 in 10 fail to stop completely for stop signs or right on red; almost nobody yields to pedestrians in crosswalks; I could go on. So will these devices be programmed to drive the way that people really drive, or in strict accordance with the law? I could see the first option being unpopular and the second exposing the manufacturer to liability.

Then there's the cyclist's perspective. How will these things be programmed to drive around cyclists? In many states it's illegal to pass a cyclist on a double yellow. Will they just wait if there's a cyclist in the road? Will they honk? That's illegal too.

Or will we see a wholesale revision of the traffic laws? One, most likely, that would further marginalize non-motorized vehicles, and even non-automated ones.
100% computer controlled means stuff like speed limits and stops signs will be a thing of the past. You don't need to be told to stop if you know of all vehicles around you and their speeds/intentions. Speed limits will be determined by the road, it's width, it's use, your vehicle, the density of traffic, as opposed to a flat speed limit determined by some city engineer that basically pulled it out of his arse. In other words: probably increased all over.

I figure if we went 100% computer controlled cars, all the laws would be re-worked for efficiency with logical circuit boards, as opposed to unruly and unreliable humans.

Driverless cars....

what's next? riderless bicycles?????

100% computer controlled means stuff like speed limits and stops signs will be a thing of the past. You don't need to be told to stop if you know of all vehicles around you and their speeds/intentions. Speed limits will be determined by the road, it's width, it's use, your vehicle, the density of traffic, as opposed to a flat speed limit determined by some city engineer that basically pulled it out of his arse. In other words: probably increased all over.

I figure if we went 100% computer controlled cars, all the laws would be re-worked for efficiency with logical circuit boards, as opposed to unruly and unreliable humans.

Rather interesting point... but how might such computer controlled vehicle know of a cyclist approaching from a side street.

While the algorithms for driving may include "speed up" commands on controlled access roads, most likely other streets will be treated as they always have been... unless we all begin wearing transponders.

It would however be quite interesting to have all the visual pollution of keep right, stop, trucks entering, etc and all the other signs removed from the roadway. For that matter, perhaps roadway stripes need not exist in the future either.

Rather interesting point... but how might such computer controlled vehicle know of a cyclist approaching from a side street.


There would still have to be some sort of sensor system to detect vehicles and objects (animate and inanimate) that might impinge on the computer-controlled car's space.

I can see, however, where speed selection could be much more dynamic, based on things like the road surface, the visibility of the sensor system, and the vehicle's ability to react to something being detected within the visibility horizon. A system like this would be able to make a much better decision on a "safe" speed than human drivers who, even with guidelines like speed limit signs and supposed intelligence, select a speed that's "too fast for conditions".

There would still have to be some sort of sensor system to detect vehicles and objects (animate and inanimate) that might impinge on the computer-controlled car's space.

I can see, however, where speed selection could be much more dynamic, based on things like the road surface, the visibility of the sensor system, and the vehicle's ability to react to something being detected within the visibility horizon. A system like this would be able to make a much better decision on a "safe" speed than human drivers who, even with guidelines like speed limit signs and supposed intelligence, select a speed that's "too fast for conditions".

Yeah that too, and it can determine easily the most minor slippage in the normal operation of the vehicle and further slow down.

It can be truly a lifesaver, however I still love driving my car. Oh well.

Let's just hope the OS for these cars isn't Microsoft.

Let's just hope the OS for these cars isn't Microsoft.


Actually MS announced a while back that they were indeed getting into doing the OS for this... sure hate to see the "blue screen of death" when the car was in charge.

Actually MS announced a while back that they were indeed getting into doing the OS for this... sure hate to see the "blue screen of death" when the car was in charge.

Heh, it would give new meaning to the error message FATAL ERROR IN KERNEL.EXE.

The only self-piloting car I want is the one from Larry Niven's short story, "Safe at Any Speed". :D

Ah yes, the car controlled by a Microsoft operating system -- the only vehicle which requires one to press "start" to shut it down or to "close all windows" if it stalls.

Self-piloting vehicles are undoubtedly the wave of the future -- it is up to us as advocates for bicyclists and pedestrians to make sure that our access and mobility rights are properly protected during the long transition.

The only self-piloting car I want is the one from Larry Niven's short story, "Safe at Any Speed". :D

Is that the story where the car "cocoons" the passenger after an accident... and releases him only after a long time?

Ah yes, the car controlled by a Microsoft operating system -- the only vehicle which requires one to press "start" to shut it down or to "close all windows" if it stalls.

Of course just because MS is writing software to try to do this doesn't mean they will be successful... MS was never able to enter the complex CAD market.

I wonder if we'll be able to upgrade to say a linux OS instead... but then again command line driving might be a real chore. ;)



Self-piloting vehicles are undoubtedly the wave of the future -- it is up to us as advocates for bicyclists and pedestrians to make sure that our access and mobility rights are properly protected during the long transition.

This is really my whole point for posting this here... indeed we do have to ensure that cyclists' rights are not infringed by this "wave of the future."

The wave of the future will involve computers, but I don't think they will be in the drivers seat, or at least it won't be a huge leap from today's car to the computer-controlled wondercar. I imagine a more HUD-type drivers aide system will be what works it's way into cars. Just like the Volvo's that have collision avoidance warnings, the Mercedes Benz S Class that can work out braking distance and has infared night vision, these are all things that will improve driving. I think it's a logical step from infared night vision to infared-based pedestrian detection, for example.

If I get a choice, I'll continue to be a driver. Microsoft can't program an operating system for desktops - why should they do one for cars?

It would however be quite interesting to have all the visual pollution of keep right, stop, trucks entering, etc and all the other signs removed from the roadway. For that matter, perhaps roadway stripes need not exist in the future either.I like this extrapolation, and agree with it. Taken further, if all occupants are busy sending text messages, surfing the web, watching movies, sleeping or whatever, then hopefully this could also result in the loss of the hideous bill boards from the landscape.

This technology should be rolled out as soon as possible. The environmental benefits from more efficient use of vehicles (cost saving), the massive reduction in collisions (cost saving) and associated reduction in liability insurance (cost saving) all lend themselves to why this should be done.

I've said it for a long time, that technology exists to prevent things like vehicle theft - but the whole auto sector and especially insurance companies have a vested interested in fender benders, vehicle theft and built in obsoleteness.

What would be the impact on the auto insurance industry (a bunch of Shylocks) if the probability of accidents falls close to zero, vehicle theft disappears and even drunks can get home safely? And don't think the economy would suffer for 1 second if the insurance sector experiences a 90% correction in it's size. Instead of a few fat cat CEO' buying luxury yachts we all get to buy super wide screen flat panel LCD tv's, that use less electricity than the conventional tv - SCORE!!!!

I'm all for the self driving car.
I don't think the robot driver will "gun it" when he sees a bike.

Remember the three laws of Robotics, from Isaac Asimov's book "I, Robot"?

Here's they are , from Wikipedia:

In science fiction, the Three Laws of Robotics are a set of three rules written by Isaac Asimov, which almost all positronic robots appearing in his fiction must obey. Introduced in his 1942 short story "Runaround," although foreshadowed in a few earlier stories, the Laws state the following:

A robot may not injure a human being or, through inaction, allow a human being to come to harm.

A robot must obey orders given to it by human beings, except where such orders would conflict with the First Law.

A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

And the link:

http://en.wikipedia.org/wiki/Three_Laws_of_Robotics

I think it would be ILLEGAL for a self driving car to be programmed to speed. What are the " 50 in a 30" SUV drivers going to do? Ram the first one they see?

I can't wait to see what happens when it goes to court.

IIRC, the self driving car will have infrared vision, so it will stop for anything in the road that gives off body heat (like a cyclist, a moose, or a deer).

Its too early to worry about it.

Undoubtedly the first self-driving cars will only be allowed on specially designated freeways -- similar to the HOV lanes now. Freeway driving is where automation would provide the biggest bang for the buck. The rules are well established, the entrances and exits controlled, and traffic more predictable -- if you take humans out of the picture. Automation would allow each car to take a position that would maximize overall traffic flow and minimize driving time. In constant communication, maybe they could even safely draft each other. Millions of vehicles all driven in synchrony could reduce gas consumption, relieve congestion, and perhaps eliminate the 45,000 annual automobile deaths.

Since bikes aren't allowed on most freeways anyways, the immediate impact is a long ways off.

This technology should be rolled out as soon as possible. The environmental benefits from more efficient use of vehicles (cost saving), the massive reduction in collisions (cost saving) and associated reduction in liability insurance (cost saving) all lend themselves to why this should be done.


On the other hand, I could see this techology leading to a dramatic increase in the number of miles that people drive, because people could use travel time for other productive uses. The environmental cost of that increased travel, from direct pollution due to increased fuel use to environmental degradation due to increased sprawl, would be substantial.

On the other hand, I could see this techology leading to a dramatic increase in the number of miles that people drive, because people could use travel time for other productive uses. The environmental cost of that increased travel, from direct pollution due to increased fuel use to environmental degradation due to increased sprawl, would be substantial.

Not with oil in the $100 region and 12 month futures of double that. Throw in the current protracted recession that looks every day like it is more likely to be a depression and increased travel might not be an option for 95% of us.

Actually MS announced a while back that they were indeed getting into doing the OS for this... sure hate to see the "blue screen of death" when the car was in charge.
As sad as it sounds, the OS for a car would be much simpler, with a higher reason to not fail.

As the industry saying (sadly) goes, "it can be patched" when it comes to desktop development software, and Microsoft loves that saying, however in stuff like embedded software for cars, no, they'd get it right the first time, or they wont do it at all.

Personally I don't think they'll follow through.


The wave of the future will involve computers, but I don't think they will be in the drivers seat, or at least it won't be a huge leap from today's car to the computer-controlled wondercar. I imagine a more HUD-type drivers aide system will be what works it's way into cars. Just like the Volvo's that have collision avoidance warnings, the Mercedes Benz S Class that can work out braking distance and has infared night vision, these are all things that will improve driving. I think it's a logical step from infared night vision to infared-based pedestrian detection, for example.

If I get a choice, I'll continue to be a driver. Microsoft can't program an operating system for desktops - why should they do one for cars?
You can put all the fancy features in a car, but you still can't keep the idiot thats behind the wheel to be less of an idiot.

However, being as I drive at my job, I always think of fun HUD ideas along with inline mirror replacements (a-pillar = mirror cams). It would be cool, but obviously we're taking our sweet time rolling out the technology.


On the other hand, I could see this techology leading to a dramatic increase in the number of miles that people drive, because people could use travel time for other productive uses. The environmental cost of that increased travel, from direct pollution due to increased fuel use to environmental degradation due to increased sprawl, would be substantial.
Assuming we reach the stage where we aren't technology fearing monkeys, we'd probably resort to other methods of energy producing too. We have a lot of maddyfishes to overcome, and they're holding us back in the power sector too because of their ignorance of those subjects too.

Is that the story where the car "cocoons" the passenger after an accident... and releases him only after a long time?

Close.

Enroute from A to B, the car gets snatched out of mid air and swallowed by a large bird-like creature. The car continues on its way, dragging the surprised roc along until, because the car's sensors were blinded by the body of the bird, it crashes. Unfortunately, the car's communications equipment is also blocked by the (now) carcass, so the driver has to wait around for several months (while being entertained and fed by the car) for the carcass to rot away before calling for a tow.

Re: autonomous-drive vehicles

The interesting thing is that GM touted a similar technology as long ago as 1939 at the World's Fair in their famous "Futurama" exhibit. It was part of the effort to sell Americans on the idea of freeways, which was supposed to "clean up crowded, dirty cities" and allow Americans to live in idyllic suburbs with "plenty of space." The automobile was the key to this bright, new future.

The freeways survived, but the promised technology that provided radio-controlled, automatic speed limiting and maintained perfect distances between automobiles never materialized.

My unprofessional guess is that autonomous-drive vehicles will not catch on, because most people prefer to control their driving in the city. It does, however, add a certain "glimmer" to the automobile, much as it did in 1939. I suspect most people would be more like Will Smith in "I, Robot" as he hated to let the auto-drive operate his vehicle in futuristic Chicago.

But gimmicks are par for the course in the high-pressure world of car sales. After all, we never see traffic jams in car advertisements, just exhilirated motorists flying down an empty highway...

"Zoom, zoom, zoom!"

;) Like in this illustration--

http://d.yimg.com/us.yimg.com/p/ap/20080106/capt.eef8e207f2154eb8bc73e12b5ed8a032.gadget_show_driverless_cars_dt103.jpg?x=400&y=266&sig=CNH1b046Zj4pEhZODKeq8Q--


It might work in certain circumstances for long-distance driving, but what if you "lost a signal" or something, like happens on your cell-phone in the mountains? Yikes...

Re: autonomous-drive vehicles

The interesting thing is that GM touted a similar technology as long ago as 1939 at the World's Fair in their famous "Futurama" exhibit. It was part of the effort to sell Americans on the idea of freeways, which was supposed to "clean up crowded, dirty cities" and allow Americans to live in idyllic suburbs with "plenty of space." The automobile was the key to this bright, new future.

The freeways survived, but the promised technology that provided radio-controlled, automatic speed limiting and maintained perfect distances between automobiles never materialized.

My unprofessional guess is that autonomous-drive vehicles will not catch on, because most people prefer to control their driving in the city. It does, however, add a certain "glimmer" to the automobile, much as it did in 1939. I suspect most people would be more like Will Smith in "I, Robot" as he hated to let the auto-drive operate his vehicle in futuristic Chicago.

But gimmicks are par for the course in the high-pressure world of car sales. After all, we never see traffic jams in car advertisements, just exhilirated motorists flying down an empty highway...

"Zoom, zoom, zoom!"

;) Like in this illustration--

http://d.yimg.com/us.yimg.com/p/ap/20080106/capt.eef8e207f2154eb8bc73e12b5ed8a032.gadget_show_driverless_cars_dt103.jpg?x=400&y=266&sig=CNH1b046Zj4pEhZODKeq8Q--


It might work in certain circumstances for long-distance driving, but what if you "lost a signal" or something, like happens on your cell-phone in the mountains? Yikes...

Well you may be right on the GM aspect of being able to deliver... but as far as the technology, it has been beta tested in several ways... unlike the 1939 false promise which technically included no way to actually deliver the "promises."

Back in the late '90s locally several auto manufactures demonstrated a system here in San Diego that allowed modified autos to get on a modified freeway, drive without human intervention, avoid obstacles, pack close together in "trains" and deliver the human passenger to the designated exit.

And a couple of years ago the DARPA project for autonomous vehicles demonstrated vehicles that were able to make their own decisions (without any needed outside signal... including GPS) and drive to a destination, without a predetermined route, while avoiding obstacles, and through a simulated city environment.

Further, there are now active sensor systems available that can allow a car to self park. There are also collision avoidance systems available that can warn drivers of obstacles the driver cannot see, or fails to notice.

The bottom line is that we are right at the cusp of all this stuff coming together into a viable vehicle that can be delivered to customers. (compare this to the days when cell phones were huge and only mounted in the trunks of cars)

The next hurdles are probably not technical... the technical roadmaps are well understood quite achievable. The next hurdles are probably legal and marketing... will such a vehicle be allowed on the streets of America... will there be strict requirements, and how will the consumer face the issues of allowing computer(s) to be in control.

As others have pointed out, computer control has long been in use in areas such as flight control. There was resistance there too by pilots. (I was a technician for such a system back in the early 80s, and pilots used to comment that they hated it).

To me the advantages are obvious... in that computer controlled cars will never drive in anger or haste, and never have opinions about who "belongs on the road" or not. And if the passengers desire to be on cell phones, or play video games... so be it... the "driver" will not be distracted.

BTW Futurama promised flying cars too... I'm still waiting.

I'm sure the advantages are real. But I'm not sure motorists will go for it, any more than cyclists are rushing out to buy electric bicycles.

But I'm arguing about the future, a very sketchy thing to argue about. If history has taught us anything, it should be "never say never."

:beer:

It might work in certain circumstances for long-distance driving, but what if you "lost a signal" or something, like happens on your cell-phone in the mountains? Yikes...
It's not reporting to a tower, but itself, the only way you'd lose a signal is if something was between you and the object you bounced off of, in which case you'd get the location of the object between you.... which is what you want anyway.


I'm sure the advantages are real. But I'm not sure motorists will go for it, any more than cyclists are rushing out to buy electric bicycles.

But I'm arguing about the future, a very sketchy thing to argue about. If history has taught us anything, it should be "never say never."

:beer:
Once we switch to electric and swap our plants for cleaner and cheaper power, I'd definitely go for it, could you imagine packing your car, going to sleep and waking up half way across the country, playing some games for the rest of the trip and you're done, at little cost to you.

Mainly because electric cars will probably ruin much of my love of driving (manual transmission and all).

I'd definitely go for it, could you imagine packing your car, going to sleep and waking up half way across the country, playing some games for the rest of the trip and you're done, at little cost to you.

I can imagine that.

Guys, I'm not a vision-buster. I told my wife last week (while our internet was down and I couldn't read these forums) that if the money was behind it, we could produce a solar-cell the size of a quarter that would power a car with small, lightweight batteries without any gasoline at all. If the money was behind it...

If the technology is implemented, it will cost plenty. It will be optional, and coveted. Those who enjoy it will pay, and they will also have the option of driving manually.

I don't forsee everyone being required to use it, and everyone who does use it will shell out big bucks.

It's not reporting to a tower, but itself, the only way you'd lose a signal is if something was between you and the object you bounced off of, in which case you'd get the location of the object between you.... which is what you want anyway.
If it's not reporting to anything but itself, how does it read its global position?

I guess it would be extremely welcome in a traffic jam. Nothing worse than wasting my time there, when I could be doing something more interesting...

...like riding a bike. :D

I can imagine that.

Guys, I'm not a vision-buster. I told my wife last week (while our internet was down and I couldn't read these forums) that if the money was behind it, we could produce a solar-cell the size of a quarter that would power a car with small, lightweight batteries without any gasoline at all. If the money was behind it...
A solar cell the size of a quarter powering your car? I'll assume that was a bit of hyperbole.

The sun rains down an average of 1000 watts per square meter. So if the entirety of a typical car was covered with 100% efficient solar cells you'd still only have about the equivalent of two hair dryers. Unfortunately, really good (i.e. expensive) solar cells have only about 15% efficiency.

Now if they could only make fusion-on-a-chip, then we'd have something.

Gosh, 8% of the way into the 21st century, and we're no closer to practical fusion than we are to practical space flight. Or flying cars. Or jet backpacks. Or domed cities. The future just isn't what it used to be.

I can imagine that.

Guys, I'm not a vision-buster. I told my wife last week (while our internet was down and I couldn't read these forums) that if the money was behind it, we could produce a solar-cell the size of a quarter that would power a car with small, lightweight batteries without any gasoline at all. If the money was behind it...

If the technology is implemented, it will cost plenty. It will be optional, and coveted. Those who enjoy it will pay, and they will also have the option of driving manually.

I don't forsee everyone being required to use it, and everyone who does use it will shell out big bucks.

Well maybe not a solarcell the size of a quarter... that would be pretty phenomenal. But at least one company has figured out a way to vastly stretch current automotive technology...

AFS Trinity Power Corporation showed off its plug-in hybrid SUV, the XH-150, at the 2008 Detroit Auto Show. The car's power train represents a refinement of the hybrid concept by employing ultra-capacitors for fast acceleration under electric power. Traditional hybrids use batteries to power low speed driving, but kick in a gas engine when high speeds or heavy acceleration is required. AFS Trinity builds on this concept by adding ultracapacitors, which, unlike batteries, are designed to quickly discharge electricity, providing power to the hybrid's motors for fast acceleration. The company's demonstration vehicle, the XH-150, also features plug-in technology, letting its battery pack get charged from the electricity grid. Plug-in hybrids, without AFS Trinity's ultracapacitors, are reputed to get 100 mpg. The company tested the XH-150 at Michelin's Laurens Proving Grounds in South Carolina, and claims impressive performance, such as 150 mpg using the EPA Combined Urban/Highway Driving Cycle, a range of 40 miles under electric power, and full range of 400 miles using both electricity and gasoline. Using both gas and electricity, the car's 0 to 60 mph time was clocked at 6.9 seconds, while under electric power it took 11.5 seconds. Its top speed is 87 mph. The company is showing off the XH-150 with the intention of licensing its ultracapacitor technology to automakers.


http://reviews.cnet.com/8301-13746_7-9849660-48.html

**********************************************************

Now the real question... what are the odds that this ultracapacitor concept gets buried by the oil/auto alliance that exists today???

Further, one has to wonder why a major auto company hasn't thought of this yet.

If it's not reporting to anything but itself, how does it read its global position?

I guess it would be extremely welcome in a traffic jam. Nothing worse than wasting my time there, when I could be doing something more interesting...

...like riding a bike. :D


I saw an engineering presentation about the DARPA autonomous drive vehicles... for location info they combined several technologies, the least of which was GPS and simple inertial navigation, but mostly it simply depended on "reading" the road as it went along...

If people like cars because of the control they have in them, if that control is taken away, they may switch to bicycles...

If it's not reporting to anything but itself, how does it read its global position?

I guess it would be extremely welcome in a traffic jam. Nothing worse than wasting my time there, when I could be doing something more interesting...

...like riding a bike. :D
It shouldn't be using a GPS system to find it's location on the road (lane usage and such), so much as what ramps/roads it needs to take, a lost signal would be like a person driving a car without a map, wandering until it gets a signal. Though I would figure it would only use GPS to do error correcting, being as the car would be able to track itself on a map after it has it's current location, and would only check the GPS every once in awhile to make sure it's accurate. Also, comparing the road to maps and distances from crossings could be cross checked for map accuracy too, so an internal map would be used like 99% of the time I'd figure.

However, you don't report to a GPS satellite, it's the other way around.

If I am hit by a driverless car do I get to then sue deep pockets like GM or better yet MS?

Also as someone who uses a PC pretty much 12 hours a day, I have yet to go a week without taking a long coffee break while the IT gurus tweak something on one of my boxes. How much are people going to like having to take their car to the mech once a week?

Will they remember to test the software in a blizzard, in an ice storm, in high winds, around bicycles? Most importantly will they remember to test it around the unpredicitability of children?

So long as we are clear that I become half owner of MS or GM in the event I or a loved one is hit by one of these, then I am all for it. Without this sort of threat hanging over them I don't see these "cut corners at any cost" and "forget functionality in favor of bells and whistles" companies doing the appropriate testing and redundencies.